On the Number of Search Queries Required for Internet Plagiarism Detection

Butakov, Sergey; Shcherbinin, Vladislav

doi:10.1109/icalt.2009.78

Cited by 6 publications

(8 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Crot, one example, utilizes a global search API from Microsoft's Bing search engine to perform this kind of selection. It uses a sliding window x words in length, thus sending to the search engine all the phrases from the document that have been formed by this sliding window and in doing so performs a very exhaustive search (Butakov & Shcherbinin, 2009).…”

Section: Related Work and Existing Solutionsmentioning

confidence: 99%

“…Child (2010) have illustrated, plugging exact phrases into a public search engine can be an effective way to locate the source of a suspicious paper, but the question of how to locate such an indicative phrase in a document remains. Butakov and Shcherbinin (2009) also indicated that if a significant part of the paper was plagiarized from the Internet, there was no need to send all possible queries to the search engine: Even as few as 10% of these can help to locate the source.…”

Section: Pdsmentioning

confidence: 99%

See 1 more Smart Citation

Protecting students' intellectual property in the web plagiarism detection process

Butakov¹,

Dyagilev²,

Tskhay³

2012

IRRODL

Self Cite

View full text Add to dashboard Cite

<p class="AbstractText">Learning management systems (LMS) play a central role in communications in online and distance education. In the digital era, with all the information now accessible at students’ fingertips, plagiarism detection services (PDS) have become a must-have part of LMS. Such integration provides a seamless experience for users, allowing PDS to check submitted digital artifacts without any noticeable effort by either professor or student. In most such systems, to compare a submitted work with possible sources on the Internet, the university transfers the student’s submission to a third-party service. Such an approach is often criticized by students, who regard this process as a violation of copyright law. To address this issue, this paper outlines an improved approach for PDS development that should allow universities to avoid such criticism. The major proposed alteration of the mainstream architecture is to move document preprocessing and search result clarification from the third-party system back to the university system. The proposed architecture changes would allow schools to submit only limited information to the third party and avoid criticism about intellectual property violation. <br /><br /></p>

show abstract

Section: Related Work and Existing Solutionsmentioning

confidence: 99%

Section: Pdsmentioning

confidence: 99%

Protecting students' intellectual property in the web plagiarism detection process

Butakov¹,

Dyagilev²,

Tskhay³

2012

IRRODL

Self Cite

View full text Add to dashboard Cite

show abstract

“…According to the legal requirements provided in the previous section these queries should not contain enough information to recover the submission but from an information retrieval point of view these queries should be enough to find similar documents on the web. Experiments indicate that even limited numbers of properly selected search queries can help to locate plagiarism sources on the web [6]. Essentially this means that the part of the PDS located on school infrastructure can prepare some queries from the key parts of the text, randomly shuffle them and send them as one large query to the external part of the PDS.…”

Section: Legal Issues Related To Plagiarism Detectionmentioning

confidence: 99%

“…Crot, one of the free PDS, uses similar brute force approach for search. It uses a sliding window of X words length, thus sending to the search engine all the phrases from the document that have been formed by this sliding window and so performing a very exhaustive search [6]. Two reports indicated that actually exhaustive search is not required to detect plagiarism.…”

mentioning

confidence: 99%

“…As Culwin & Child have indicated, the exact phrase search technique with use of public search engine can be effective to locate the source [10]. Crot authors also indicated that if a significant part of the paper was plagiarized from the Internet there is no need to send all possible queries to the search engine: even as low as 10% of these queries can help to locate the source [6]. This reduction can be very helpful for IP protection and will be discussed further in section 4.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Architecture of Plagiarism Detection Service that Does Not Violate Intellectual Property of the Student

Butakov

Barber²,

Diagilev

et al. 2011

Proceedings of the 8th International Workshop on Security in Information Systems

View full text Add to dashboard Cite

Plagiarism detection services (PDS) have become a vital part of Learning Management Systems (LMS). Commercial or non-commercial PDS can be easily attached to the most popular LMS these days. In most such systems, to compare a submitted work with all possible sources on the Internet a university has to transfer the student submission to the third party. Such an approach is often criticized by students who may see a violation of copyright law in this process. This paper outlines an improved approach for PDS development that should allow universities to avoid such criticism. The major proposed alteration of the mainstream architecture of the improved PDS is a move of document preprocessing and search result clarification from the server side to the client side. Such a split allows users to submit only limited information to the third party, and to do so in a way that will not make it possible to fully recover the submitted work but will allow the PDS to maintain the same search quality.

show abstract